End of training

README.md

@@ -14,9 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Whisper-small-new
 
-This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0016
+- Loss: 0.0020
 
 ## Model description
 
@@ -49,16 +49,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.1234        | 1.0   | 137  | 0.0824          |
-| 0.0749        | 2.0   | 274  | 0.0623          |
-| 0.0378        | 3.0   | 411  | 0.0587          |
-| 0.049         | 4.0   | 548  | 0.0214          |
-| 0.0213        | 5.0   | 685  | 0.0157          |
-| 0.01          | 6.0   | 822  | 0.0097          |
-| 0.0077        | 7.0   | 959  | 0.0052          |
-| 0.0059        | 8.0   | 1096 | 0.0025          |
-| 0.0021        | 9.0   | 1233 | 0.0017          |
-| 0.0019        | 10.0  | 1370 | 0.0016          |
+| 0.1302        | 1.0   | 137  | 0.0964          |
+| 0.0897        | 2.0   | 274  | 0.0545          |
+| 0.0691        | 3.0   | 411  | 0.0396          |
+| 0.0482        | 4.0   | 548  | 0.0279          |
+| 0.0374        | 5.0   | 685  | 0.0174          |
+| 0.0189        | 6.0   | 822  | 0.0096          |
+| 0.0108        | 7.0   | 959  | 0.0063          |
+| 0.004         | 8.0   | 1096 | 0.0039          |
+| 0.0039        | 9.0   | 1233 | 0.0024          |
+| 0.0026        | 10.0  | 1370 | 0.0020          |
 
 
 ### Framework versions

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3be42fda9d1cbc9320d4744425ed1f9cd861210aa307949b916e7665a179509f
+oid sha256:8aea2e642780d477876c5360586f438351a721486a846c3783bf84083ded947e
 size 14176064

runs/Jun17_03-32-19_fc8d453ac50a/events.out.tfevents.1718595151.fc8d453ac50a.279.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:215c895b696a4967ffb51dcef31fc61c2418ba0beb51f37310e857b86d94df53
+size 20613

trainer_state.json

@@ -10,470 +10,470 @@
   "log_history": [
     {
       "epoch": 0.18248175182481752,
-      "grad_norm": 2.397657632827759,
-      "learning_rate": 0.00048,
-      "loss": 4.0044,
+      "grad_norm": 23.73174476623535,
+      "learning_rate": 0.00046,
+      "loss": 4.0301,
       "step": 25
     },
     {
       "epoch": 0.36496350364963503,
-      "grad_norm": 0.8800250291824341,
-      "learning_rate": 0.00098,
-      "loss": 0.9921,
+      "grad_norm": 2.086200714111328,
+      "learning_rate": 0.00096,
+      "loss": 1.0569,
       "step": 50
     },
     {
       "epoch": 0.5474452554744526,
-      "grad_norm": 1.0678032636642456,
+      "grad_norm": 0.4636005461215973,
       "learning_rate": 0.0009825757575757576,
-      "loss": 0.4551,
+      "loss": 0.4231,
       "step": 75
     },
     {
       "epoch": 0.7299270072992701,
-      "grad_norm": 8.535021781921387,
+      "grad_norm": 0.905696451663971,
       "learning_rate": 0.0009636363636363637,
-      "loss": 0.148,
+      "loss": 0.1587,
       "step": 100
     },
     {
       "epoch": 0.9124087591240876,
-      "grad_norm": 0.5009177327156067,
+      "grad_norm": 0.7420445084571838,
       "learning_rate": 0.0009446969696969697,
-      "loss": 0.1234,
+      "loss": 0.1302,
       "step": 125
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.08237148076295853,
-      "eval_runtime": 191.2601,
-      "eval_samples_per_second": 4.277,
-      "eval_steps_per_second": 0.716,
+      "eval_loss": 0.09637967497110367,
+      "eval_runtime": 192.8711,
+      "eval_samples_per_second": 4.241,
+      "eval_steps_per_second": 0.71,
       "step": 137
     },
     {
       "epoch": 1.094890510948905,
-      "grad_norm": 0.6111318469047546,
+      "grad_norm": 0.5315635204315186,
       "learning_rate": 0.0009257575757575758,
-      "loss": 0.1034,
+      "loss": 0.1168,
       "step": 150
     },
     {
       "epoch": 1.2773722627737225,
-      "grad_norm": 0.47145724296569824,
+      "grad_norm": 2.168321132659912,
       "learning_rate": 0.0009068181818181819,
-      "loss": 0.0704,
+      "loss": 0.0957,
       "step": 175
     },
     {
       "epoch": 1.4598540145985401,
-      "grad_norm": 0.3196675777435303,
+      "grad_norm": 1.4148447513580322,
       "learning_rate": 0.000887878787878788,
-      "loss": 0.0841,
+      "loss": 0.0819,
       "step": 200
     },
     {
       "epoch": 1.6423357664233578,
-      "grad_norm": 1.194128394126892,
+      "grad_norm": 1.08322274684906,
       "learning_rate": 0.0008689393939393939,
-      "loss": 0.1131,
+      "loss": 0.0967,
       "step": 225
     },
     {
       "epoch": 1.8248175182481752,
-      "grad_norm": 0.5795499086380005,
+      "grad_norm": 0.34484872221946716,
       "learning_rate": 0.00085,
-      "loss": 0.0749,
+      "loss": 0.0897,
       "step": 250
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.06233128905296326,
-      "eval_runtime": 192.0678,
-      "eval_samples_per_second": 4.259,
-      "eval_steps_per_second": 0.713,
+      "eval_loss": 0.05452003329992294,
+      "eval_runtime": 193.1249,
+      "eval_samples_per_second": 4.236,
+      "eval_steps_per_second": 0.709,
       "step": 274
     },
     {
       "epoch": 2.0072992700729926,
-      "grad_norm": 1.2165534496307373,
+      "grad_norm": 0.4588640034198761,
       "learning_rate": 0.0008310606060606061,
-      "loss": 0.1015,
+      "loss": 0.078,
       "step": 275
     },
     {
       "epoch": 2.18978102189781,
-      "grad_norm": 0.4674736261367798,
+      "grad_norm": 0.9430391788482666,
       "learning_rate": 0.0008121212121212122,
-      "loss": 0.0487,
+      "loss": 0.047,
       "step": 300
     },
     {
       "epoch": 2.372262773722628,
-      "grad_norm": 0.28624579310417175,
+      "grad_norm": 1.2605994939804077,
       "learning_rate": 0.0007931818181818182,
-      "loss": 0.0979,
+      "loss": 0.0733,
       "step": 325
     },
     {
       "epoch": 2.554744525547445,
-      "grad_norm": 0.5534139275550842,
+      "grad_norm": 0.4879356026649475,
       "learning_rate": 0.0007742424242424244,
-      "loss": 0.0503,
+      "loss": 0.0595,
       "step": 350
     },
     {
       "epoch": 2.7372262773722627,
-      "grad_norm": 0.959182620048523,
+      "grad_norm": 0.7468044757843018,
       "learning_rate": 0.0007553030303030303,
-      "loss": 0.0513,
+      "loss": 0.0581,
       "step": 375
     },
     {
       "epoch": 2.9197080291970803,
-      "grad_norm": 0.3576785624027252,
+      "grad_norm": 1.2231853008270264,
       "learning_rate": 0.0007363636363636363,
-      "loss": 0.0378,
+      "loss": 0.0691,
       "step": 400
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.05872492492198944,
-      "eval_runtime": 187.3215,
-      "eval_samples_per_second": 4.367,
-      "eval_steps_per_second": 0.731,
+      "eval_loss": 0.039563629776239395,
+      "eval_runtime": 191.6012,
+      "eval_samples_per_second": 4.269,
+      "eval_steps_per_second": 0.715,
       "step": 411
     },
     {
       "epoch": 3.102189781021898,
-      "grad_norm": 0.5636082291603088,
+      "grad_norm": 0.3177865445613861,
       "learning_rate": 0.0007174242424242424,
-      "loss": 0.0642,
+      "loss": 0.0421,
       "step": 425
     },
     {
       "epoch": 3.2846715328467155,
-      "grad_norm": 0.3895982503890991,
+      "grad_norm": 1.0183290243148804,
       "learning_rate": 0.0006984848484848485,
-      "loss": 0.0308,
+      "loss": 0.0485,
       "step": 450
     },
     {
       "epoch": 3.4671532846715327,
-      "grad_norm": 0.28794267773628235,
+      "grad_norm": 0.361370712518692,
       "learning_rate": 0.0006795454545454546,
-      "loss": 0.0374,
+      "loss": 0.0369,
       "step": 475
     },
     {
       "epoch": 3.6496350364963503,
-      "grad_norm": 0.526278018951416,
+      "grad_norm": 1.0689235925674438,
       "learning_rate": 0.0006606060606060606,
-      "loss": 0.0331,
+      "loss": 0.0456,
       "step": 500
     },
     {
       "epoch": 3.832116788321168,
-      "grad_norm": 0.30551332235336304,
+      "grad_norm": 0.29606688022613525,
       "learning_rate": 0.0006416666666666667,
-      "loss": 0.049,
+      "loss": 0.0482,
       "step": 525
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.021386895328760147,
-      "eval_runtime": 189.2731,
-      "eval_samples_per_second": 4.322,
-      "eval_steps_per_second": 0.724,
+      "eval_loss": 0.027906004339456558,
+      "eval_runtime": 190.9612,
+      "eval_samples_per_second": 4.284,
+      "eval_steps_per_second": 0.717,
       "step": 548
     },
     {
       "epoch": 4.014598540145985,
-      "grad_norm": 0.11820922046899796,
+      "grad_norm": 0.1826704740524292,
       "learning_rate": 0.0006227272727272727,
-      "loss": 0.0299,
+      "loss": 0.0451,
       "step": 550
     },
     {
       "epoch": 4.197080291970803,
-      "grad_norm": 0.2699453830718994,
+      "grad_norm": 0.1476697325706482,
       "learning_rate": 0.0006037878787878788,
-      "loss": 0.025,
+      "loss": 0.0324,
       "step": 575
     },
     {
       "epoch": 4.37956204379562,
-      "grad_norm": 0.18448378145694733,
+      "grad_norm": 0.37707987427711487,
       "learning_rate": 0.0005848484848484848,
-      "loss": 0.0242,
+      "loss": 0.0277,
       "step": 600
     },
     {
       "epoch": 4.562043795620438,
-      "grad_norm": 0.09863670915365219,
+      "grad_norm": 0.18928465247154236,
       "learning_rate": 0.0005659090909090909,
-      "loss": 0.0188,
+      "loss": 0.0331,
       "step": 625
     },
     {
       "epoch": 4.744525547445256,
-      "grad_norm": 0.45231935381889343,
+      "grad_norm": 0.10301584750413895,
       "learning_rate": 0.000546969696969697,
-      "loss": 0.0301,
+      "loss": 0.0189,
       "step": 650
     },
     {
       "epoch": 4.927007299270073,
-      "grad_norm": 0.33688464760780334,
+      "grad_norm": 0.5264261960983276,
       "learning_rate": 0.0005280303030303031,
-      "loss": 0.0213,
+      "loss": 0.0374,
       "step": 675
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.01573045551776886,
-      "eval_runtime": 193.8274,
-      "eval_samples_per_second": 4.22,
-      "eval_steps_per_second": 0.707,
+      "eval_loss": 0.017436422407627106,
+      "eval_runtime": 189.4577,
+      "eval_samples_per_second": 4.318,
+      "eval_steps_per_second": 0.723,
       "step": 685
     },
     {
       "epoch": 5.109489051094891,
-      "grad_norm": 0.13048557937145233,
+      "grad_norm": 0.3798004984855652,
       "learning_rate": 0.000509090909090909,
-      "loss": 0.0148,
+      "loss": 0.0166,
       "step": 700
     },
     {
       "epoch": 5.291970802919708,
-      "grad_norm": 0.16118614375591278,
+      "grad_norm": 0.22640006244182587,
       "learning_rate": 0.0004901515151515152,
-      "loss": 0.0152,
+      "loss": 0.0224,
       "step": 725
     },
     {
       "epoch": 5.474452554744525,
-      "grad_norm": 0.21335643529891968,
+      "grad_norm": 0.2977535128593445,
       "learning_rate": 0.0004712121212121212,
-      "loss": 0.0108,
+      "loss": 0.0178,
       "step": 750
     },
     {
       "epoch": 5.656934306569343,
-      "grad_norm": 0.41349828243255615,
+      "grad_norm": 0.046165354549884796,
       "learning_rate": 0.00045227272727272727,
-      "loss": 0.0132,
+      "loss": 0.0188,
       "step": 775
     },
     {
       "epoch": 5.839416058394161,
-      "grad_norm": 0.2152714878320694,
+      "grad_norm": 0.20036080479621887,
       "learning_rate": 0.00043333333333333337,
-      "loss": 0.01,
+      "loss": 0.0189,
       "step": 800
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.009678781032562256,
-      "eval_runtime": 190.6043,
-      "eval_samples_per_second": 4.292,
+      "eval_loss": 0.009602065198123455,
+      "eval_runtime": 190.5013,
+      "eval_samples_per_second": 4.294,
       "eval_steps_per_second": 0.719,
       "step": 822
     },
     {
       "epoch": 6.021897810218978,
-      "grad_norm": 0.21781329810619354,
+      "grad_norm": 0.009559686295688152,
       "learning_rate": 0.00041439393939393936,
-      "loss": 0.0176,
+      "loss": 0.0112,
       "step": 825
     },
     {
       "epoch": 6.204379562043796,
-      "grad_norm": 0.08809634298086166,
+      "grad_norm": 0.17128558456897736,
       "learning_rate": 0.00039545454545454546,
-      "loss": 0.0093,
+      "loss": 0.0115,
       "step": 850
     },
     {
       "epoch": 6.386861313868613,
-      "grad_norm": 0.08477772027254105,
+      "grad_norm": 0.12201745808124542,
       "learning_rate": 0.0003765151515151515,
-      "loss": 0.0073,
+      "loss": 0.0099,
       "step": 875
     },
     {
       "epoch": 6.569343065693431,
-      "grad_norm": 0.1276165395975113,
+      "grad_norm": 0.221001997590065,
       "learning_rate": 0.0003575757575757576,
-      "loss": 0.0081,
+      "loss": 0.0062,
       "step": 900
     },
     {
       "epoch": 6.751824817518248,
-      "grad_norm": 0.24087348580360413,
+      "grad_norm": 0.1375964730978012,
       "learning_rate": 0.00033863636363636366,
-      "loss": 0.0083,
+      "loss": 0.0104,
       "step": 925
     },
     {
       "epoch": 6.934306569343065,
-      "grad_norm": 0.08206004649400711,
+      "grad_norm": 0.18805921077728271,
       "learning_rate": 0.0003196969696969697,
-      "loss": 0.0077,
+      "loss": 0.0108,
       "step": 950
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.00517745828256011,
-      "eval_runtime": 189.453,
-      "eval_samples_per_second": 4.318,
-      "eval_steps_per_second": 0.723,
+      "eval_loss": 0.006272537633776665,
+      "eval_runtime": 189.9453,
+      "eval_samples_per_second": 4.307,
+      "eval_steps_per_second": 0.721,
       "step": 959
     },
     {
       "epoch": 7.116788321167883,
-      "grad_norm": 0.0274250078946352,
+      "grad_norm": 0.04299464076757431,
       "learning_rate": 0.0003007575757575758,
-      "loss": 0.0064,
+      "loss": 0.0065,
       "step": 975
     },
     {
       "epoch": 7.299270072992701,
-      "grad_norm": 0.036731522530317307,
+      "grad_norm": 0.19402629137039185,
       "learning_rate": 0.0002818181818181818,
-      "loss": 0.004,
+      "loss": 0.0089,
       "step": 1000
     },
     {
       "epoch": 7.481751824817518,
-      "grad_norm": 0.14000241458415985,
+      "grad_norm": 0.1514054536819458,
       "learning_rate": 0.0002628787878787879,
-      "loss": 0.0041,
+      "loss": 0.007,
       "step": 1025
     },
     {
       "epoch": 7.664233576642336,
-      "grad_norm": 0.10076174885034561,
+      "grad_norm": 0.011143018491566181,
       "learning_rate": 0.00024393939393939392,
-      "loss": 0.0032,
+      "loss": 0.0074,
       "step": 1050
     },
     {
       "epoch": 7.846715328467154,
-      "grad_norm": 0.2230420559644699,
+      "grad_norm": 0.037931449711322784,
       "learning_rate": 0.00022500000000000002,
-      "loss": 0.0059,
+      "loss": 0.004,
       "step": 1075
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.00246231397613883,
-      "eval_runtime": 186.101,
-      "eval_samples_per_second": 4.395,
-      "eval_steps_per_second": 0.736,
+      "eval_loss": 0.003884958801791072,
+      "eval_runtime": 190.9247,
+      "eval_samples_per_second": 4.284,
+      "eval_steps_per_second": 0.718,
       "step": 1096
     },
     {
       "epoch": 8.02919708029197,
-      "grad_norm": 0.012392624281346798,
+      "grad_norm": 0.09535694122314453,
       "learning_rate": 0.00020606060606060607,
-      "loss": 0.004,
+      "loss": 0.0047,
       "step": 1100
     },
     {
       "epoch": 8.211678832116789,
-      "grad_norm": 0.013894147239625454,
+      "grad_norm": 0.1803082972764969,
       "learning_rate": 0.00018712121212121214,
-      "loss": 0.003,
+      "loss": 0.0047,
       "step": 1125
     },
     {
       "epoch": 8.394160583941606,
-      "grad_norm": 0.043322790414094925,
+      "grad_norm": 0.07427278906106949,
       "learning_rate": 0.0001681818181818182,
-      "loss": 0.0029,
+      "loss": 0.0035,
       "step": 1150
     },
     {
       "epoch": 8.576642335766424,
-      "grad_norm": 0.04905259236693382,
+      "grad_norm": 0.07938549667596817,
       "learning_rate": 0.00014924242424242424,
-      "loss": 0.0025,
+      "loss": 0.0038,
       "step": 1175
     },
     {
       "epoch": 8.75912408759124,
-      "grad_norm": 0.16631004214286804,
+      "grad_norm": 0.047964416444301605,
       "learning_rate": 0.0001303030303030303,
-      "loss": 0.0023,
+      "loss": 0.0035,
       "step": 1200
     },
     {
       "epoch": 8.941605839416058,
-      "grad_norm": 0.048456668853759766,
+      "grad_norm": 0.03935278207063675,
       "learning_rate": 0.00011136363636363636,
-      "loss": 0.0021,
+      "loss": 0.0039,
       "step": 1225
     },
     {
       "epoch": 9.0,
-      "eval_loss": 0.0017340199556201696,
-      "eval_runtime": 189.6437,
-      "eval_samples_per_second": 4.313,
-      "eval_steps_per_second": 0.722,
+      "eval_loss": 0.002366352593526244,
+      "eval_runtime": 191.3767,
+      "eval_samples_per_second": 4.274,
+      "eval_steps_per_second": 0.716,
       "step": 1233
     },
     {
       "epoch": 9.124087591240876,
-      "grad_norm": 0.02239864319562912,
+      "grad_norm": 0.06493715196847916,
       "learning_rate": 9.242424242424242e-05,
-      "loss": 0.0021,
+      "loss": 0.0025,
       "step": 1250
     },
     {
       "epoch": 9.306569343065693,
-      "grad_norm": 0.02109498344361782,
+      "grad_norm": 0.02973032370209694,
       "learning_rate": 7.348484848484849e-05,
-      "loss": 0.0014,
+      "loss": 0.0031,
       "step": 1275
     },
     {
       "epoch": 9.489051094890511,
-      "grad_norm": 0.013567096553742886,
+      "grad_norm": 0.03156217932701111,
       "learning_rate": 5.4545454545454546e-05,
-      "loss": 0.0017,
+      "loss": 0.0013,
       "step": 1300
     },
     {
       "epoch": 9.671532846715328,
-      "grad_norm": 0.04844328388571739,
+      "grad_norm": 0.01587042771279812,
       "learning_rate": 3.560606060606061e-05,
-      "loss": 0.0015,
+      "loss": 0.0036,
       "step": 1325
     },
     {
       "epoch": 9.854014598540147,
-      "grad_norm": 0.04474401846528053,
+      "grad_norm": 0.02338283136487007,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.0019,
+      "loss": 0.0026,
       "step": 1350
     },
     {
       "epoch": 10.0,
-      "eval_loss": 0.0015559077728539705,
-      "eval_runtime": 190.3953,
-      "eval_samples_per_second": 4.296,
-      "eval_steps_per_second": 0.72,
+      "eval_loss": 0.001961252186447382,
+      "eval_runtime": 191.871,
+      "eval_samples_per_second": 4.263,
+      "eval_steps_per_second": 0.714,
       "step": 1370
     },
     {
       "epoch": 10.0,
       "step": 1370,
       "total_flos": 2.4023145037824e+18,
-      "train_loss": 0.129441974483376,
-      "train_runtime": 4316.699,
-      "train_samples_per_second": 1.895,
-      "train_steps_per_second": 0.317
+      "train_loss": 0.1333546004319278,
+      "train_runtime": 4360.7512,
+      "train_samples_per_second": 1.876,
+      "train_steps_per_second": 0.314
     }
   ],
   "logging_steps": 25,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:189e4c51ffcb162c7ba0bceace7408f517b914cd5ef169ead0ffa5f96c9ca66d
+oid sha256:d93124ee5a88c54da280f781db2f080bede384c13fcbf6a518df8da3f51d7783
 size 5304